The primary purpose of the proposed research is to assess the generality of treatment effects (in terms of both child and parent behavior) that occurs when parents are trained to modify the behavior problems of their young children. Multiple outcome measures (i.e., observational data by independent observers in three settings, parent recorded data, parent completed questionnaires) will be collected. Four types of generality will be assessed: temporal, setting, behavioral, and sibling. A second purpose is to examine the effects of knowledge of social learning principles in facilitating generality. A third purpose is to examine the effects of a self-control/fading versus no self-control/fading treatment procedure on temporal generality.